Pickled vegetable pickling wastewater treatment system

Abstract

The present invention provides a pickled vegetable pickling wastewater treatment system, and relates to the technical field of wastewater treatment, the pickled vegetable pickling wastewater treatment system comprises the following modules: an equidistant layering module for equidistantly dividing a pickled vegetable pickling wastewater pool according to the vertical depth, and allowing the wastewater to stand to generate a plurality of continuous wastewater levels; the dynamic threshold value partitioning module is used for counting the total number of independent electrolytic layers; the multi-source coupling preprocessing module is used for collecting relevant characteristic data directly associated with the COD purification effect; and the index-driven closed-loop optimization module is used for calculating an electrolytic purification prediction index, setting an electrolytic purification index threshold value and judging whether the initial COD concentration span threshold value needs to be corrected or not. According to the invention, the independent electrolysis layer can be automatically judged and inserted according to the concentration difference of the adjacent layers, so that differential electrolysis is implemented for different concentration intervals, and the processing precision is improved. Static uniform treatment is changed into dynamic partition optimization, the treatment efficiency and the effluent quality stability are remarkably improved, and the energy consumption cost can be reduced while the purification requirement can be met.

Description

Technical Field

[0001] This invention relates to the field of wastewater treatment technology, specifically to a treatment system for wastewater from pickling vegetables. Background Technology

[0002] In the treatment of wastewater from pickling and fermenting vegetables, significant vertical concentration stratification occurs due to the long-term stagnant state of the wastewater in the pool. The pollutant load varies greatly at different depths. Therefore, by dynamically identifying concentration abrupt change ranges and setting up independent electrolysis layers, optimizing electrode arrangement and reaction stages, the mass transfer efficiency and current utilization rate of the electrochemical reaction are improved. This allows for the targeted degradation of high-salt, high-organic-load, and characteristic pollutants (such as nitrite). Multi-stage electrolysis layers enable stepwise oxidation of organic matter, avoiding electrode passivation and reducing energy consumption. Simultaneously, predicting the final purification level is also crucial, as its predictive accuracy provides a basis for process control decisions, mitigating the risk of exceeding emission standards and controlling treatment costs.

[0003] In the prior art, CN112010474A discloses a micro-electrolysis wastewater treatment system and method. This technology includes: introducing pre-adjusted pH wastewater to the bottom of a micro-electrolysis tank; the wastewater in the micro-electrolysis tank being introduced into an equalization tank through a first overflow port at the top of the micro-electrolysis tank; detecting that the pH value of the wastewater in the micro-electrolysis tank has risen to a set pH value; then, re-introducing the wastewater from the micro-electrolysis tank into the bottom of the micro-electrolysis tank to adjust the pH value of the wastewater in the equalization tank to the pre-adjusted pH value; and stopping the re-introduction of the wastewater from the micro-electrolysis tank into the bottom of the micro-electrolysis tank after reaching the pre-adjusted pH value; and performing a micro-electrolysis reaction on the wastewater using iron-carbon micro-electrolysis packing material suspended inside the micro-electrolysis tank. This solution improves the treatment effect on wastewater containing high concentrations of organic pollutants and avoids the caking phenomenon of the iron-carbon micro-electrolysis packing material.

[0004] However, the aforementioned existing technologies treat the wastewater tank as a homogeneous mixture for overall treatment, which makes it difficult to effectively identify and address significant differences in COD concentration at vertical depths. This can easily lead to local overtreatment or undertreatment during the treatment process. Furthermore, these technologies rely on fixed experience or static parameter settings and lack dynamic optimization and adjustment mechanisms for key parameters such as the number of electrolysis layers and current density. Consequently, they have poor adaptability to water quality fluctuations and unstable treatment results. At the same time, conventional solutions often exhibit high energy consumption, low purification efficiency, and difficulty in consistently meeting effluent quality standards when treating pickled vegetable wastewater with uneven concentration distribution and complex composition.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a treatment system for pickled vegetable wastewater to solve the problems mentioned in the background section. This invention can automatically determine and insert independent electrolysis layers based on the concentration difference between adjacent layers, thereby implementing differentiated electrolysis for different concentration ranges and improving treatment accuracy. It achieves a shift from static uniform treatment to dynamic zonal optimization, significantly improving treatment efficiency and effluent quality stability, and reducing energy costs while meeting purification requirements.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A system for treating wastewater from pickling vegetables includes the following modules:

[0009] Equidistant stratification module: The pickling wastewater pool is divided into equal intervals according to vertical depth. After the wastewater is allowed to settle, N continuous wastewater levels are generated. The depth range of each level is [0.3-0.5] meters. Wastewater samples are collected from each level simultaneously, and the chemical oxygen demand (COD) concentration of each level is determined by the standard potassium dichromate method to form a depth distribution curve of COD concentration.

[0010] Dynamic threshold partitioning module: Set an initial COD concentration span threshold, calculate the COD concentration difference between adjacent layers from the bottom to the top; when the COD concentration difference between adjacent layers is not less than the initial COD concentration span threshold, an independent electrolysis layer is inserted; when the COD concentration difference between adjacent layers is less than the initial COD concentration span threshold, they are merged into the same electrolysis zone; after traversing all layers, count the total number M of independent electrolysis layers, and record the depth range covered by each electrolysis layer.

[0011] Multi-source coupling preprocessing module: Collects relevant feature data directly related to COD purification effect. The relevant feature data includes electrolysis time, current density, water pollution control agent concentration, conductivity and wastewater temperature. All relevant feature data are normalized preprocessed, and missing values ​​are filled with the mean of a sliding window.

[0012] The index-driven closed-loop optimization module calculates the electrolytic purification prediction index based on the preprocessed relevant feature data collected by the multi-source coupling preprocessing module, sets the electrolytic purification index threshold, compares the electrolytic purification index threshold with the electrolytic purification prediction index, and determines whether the initial COD concentration span threshold needs to be corrected.

[0013] Furthermore, when performing equidistant division in the equidistant stratification module, the minimum number of strata is determined by calculating the pool depth divided by the lower limit of the single-layer depth (0.3 meters) and rounding up, based on the total depth of the wastewater pool and the preset single-layer depth range. At the same time, divide by the upper limit of single-layer depth of 0.5 meters and round down to determine the maximum number of layers. Finally, the number of layers was selected between the minimum and maximum values. When collecting wastewater samples from each layer simultaneously, the chemical oxygen demand (COD) concentration was measured multiple times using the standard potassium dichromate method. The average value was then used to generate a high-precision COD concentration depth distribution curve.

[0014] Furthermore, a historical wastewater treatment database was established to obtain the number of historical independent electrolytic layers used in wastewater treatment ponds of the same depth. ;

[0015] like The total number of wastewater levels N at this point is related to the number of historical independent electrolysis layers. Same, that is ;

[0016] like Then the total number of wastewater levels N is the minimum value of the total number of levels. Same, that is ;

[0017] like The total number of wastewater levels N is the maximum value of the total number of levels. Same, that is .

[0018] Furthermore, in the dynamic threshold partitioning module, the formula for calculating the total number of electrolytic layers M is:

[0019]

[0020] in:

[0021] This represents the total number of wastewater treatment levels.

[0022] This is the numbering of the wastewater level, and The value range is [1, N-1];

[0023] For the first Layer and First COD concentration difference between layers;

[0024] This represents the threshold value for the initial COD concentration range;

[0025] and For indicator functions, when the condition The value is 1 when the condition is met, and 0 otherwise. It is used to quantify whether the difference in COD concentration between adjacent levels reaches the initial COD concentration span threshold. If this condition is not met, adjacent wastewater layers merge into the same electrolysis zone, and the M value does not increase.

[0026] Furthermore, the first Layer and First COD concentration difference between layers The calculation method is as follows:

[0027]

[0028] in:

[0029] For the first COD concentration of wastewater in the layer;

[0030] For the first COD concentration of wastewater in the layer.

[0031] Furthermore, in the multi-source coupling preprocessing module, the relevant feature data undergoes unified dimension conversion and standardization processing, and the maximum-minimum normalization method is used to map each feature value to the [0,1] interval to eliminate the influence of magnitude differences on subsequent calculations; for missing values ​​that may occur during data acquisition, a sliding window mechanism is established based on time series or spatial adjacency, and the missing values ​​are filled by calculating the arithmetic mean of the effective data within the window, and the window size is dynamically adjusted to 3-5 data points;

[0032] The system performs threshold filtering on outliers, removing those that exceed the historical data mean by ±3 times the standard deviation, and initiating a redundant data replacement process.

[0033] Furthermore, the redundant data replacement process includes:

[0034] The compensation mechanism activated when outliers or missing values ​​are identified is as follows: Relevant redundant datasets of the same related feature data are collected at adjacent time points, spatial locations, or backup monitoring nodes. By comparing the spatiotemporal correlation between outlier data points and redundant datasets, valid relevant redundant data with high spatial correlation are selected as replacement candidates. The condition for high spatial correlation is: relevant feature data within a radial distance of ≤0.2 meters on the same layer.

[0035] If multiple valid candidate values ​​exist in the relevant feature redundancy dataset, a confidence-weighted fusion algorithm is used to generate the optimal replacement value. After the replacement is completed, the system immediately performs cross-parameter logical verification. If the verification fails, a second replacement is triggered. All replacement operations are marked with timestamps, original values, replacement values, and decision paths, and written to the audit log.

[0036] Furthermore, the formula for normalizing the relevant feature data in the multi-source coupling preprocessing module is as follows:

[0037]

[0038] in:

[0039] For the various relevant feature data collected;

[0040] For normalized relevant feature data;

[0041] This refers to the minimum baseline value for each relevant feature data; This represents the maximum baseline value for each relevant feature data;

[0042] During the normalization process, a corresponding static benchmark value is selected for each relevant feature data. The static benchmark value is the representative average value of wastewater after settling in the historical database. A dynamic verification mechanism for the benchmark value is established, and the benchmark value is updated periodically based on the distribution characteristics of historical data. The normalized relevant feature data are superimposed input variables.

[0043] Furthermore, in the index-driven closed-loop optimization module, the calculation formula for the electrolysis purification prediction index is as follows:

[0044]

[0045] in:

[0046] This is a predicted index for electrolysis purification.

[0047] The electrolysis time is the electrolysis time value after maximum-min normalization, ranging from [0,1].

[0048] is the current density, the current density value after maximum-min normalization, ranging from [0,1].

[0049] The concentration of water pollution control agent is the agent concentration value after maximum-minimum normalization, ranging from [0,1].

[0050] represents the conductivity, a conductivity value after maximum-minimum normalization, ranging from [0,1].

[0051] The wastewater temperature is the temperature value after maximum-minimum normalization, ranging from [0,1].

[0052] These are the electrolysis time (t) and current density, respectively. Concentration of water pollution control agents Electrical conductivity The weighting coefficients for wastewater temperature K and the total number of electrolytic layers M are determined, and each weighting coefficient satisfies the following relationship: ,and .

[0053] Furthermore, the threshold for the electrolytic purification index is set as follows: The calculated electrolysis purification prediction index With the threshold of the electrolytic purification index Perform real-time comparison:

[0054] like If the total number of electrolytic layers M and the corresponding relevant characteristic data meet the purification requirements, the process will directly enter the execution stage.

[0055] like This triggers an adaptive correction mechanism—the system dynamically reduces the COD concentration range threshold. Based on the updated COD concentration span threshold The hierarchical logic in the dynamic threshold partitioning module is re-executed to generate a larger total number of electrolysis layers. At the same time, the relevant feature data of each electrolysis layer are adjusted in conjunction with the update parameters. Then, the electrolysis purification prediction index under the updated parameters is recalculated and compared with the electrolysis purification index threshold for the second time until the electrolysis purification index threshold requirement is reached or exceeded.

[0056] Compared with the prior art, the beneficial effects of the present invention are:

[0057] This invention achieves refined identification and zoning of COD concentration differences in the vertical direction of wastewater ponds through equidistant stratification and dynamic threshold zoning. It can automatically determine and insert independent electrolysis layers based on the concentration difference between adjacent layers, thereby implementing differentiated electrolysis for different concentration ranges and improving treatment accuracy. Simultaneously, it introduces multi-source coupled pretreatment and electrolysis purification prediction index calculation, realizing a shift from static uniform treatment to dynamic zoning optimization, significantly improving treatment efficiency and effluent quality stability. This overcomes the problems of large fluctuations in treatment effect and poor adaptability of traditional methods when dealing with pickled vegetable wastewater with uneven concentration distribution, and can reduce energy consumption costs while meeting purification requirements. The system can integrate high-efficiency activated carbon adsorption technology or water pollution control agents to further enhance purification efficiency. Attached Figure Description

[0058] Figure 1 This is a schematic diagram of the wastewater treatment system for pickled vegetables according to the present invention.

[0059] Figure 2 This is a flowchart illustrating the operation of the pickled vegetable wastewater treatment system of the present invention. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0061] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0062] Example:

[0063] Please see Figures 1-2 The present invention provides the following technical solutions:

[0064] A system for treating wastewater from pickling vegetables includes the following modules:

[0065] Equidistant layered module: Based on the vertical depth of the wastewater tank, precise measurement technology is used to evenly divide the wastewater tank into multiple continuous layers, ensuring that the spacing between each layer is strictly consistent; after the wastewater has been left to stand for a period of time, it naturally forms N continuous wastewater layers. This standing process helps the wastewater in each layer to be stably separated and reduces mixing interference; the depth of each wastewater layer is controlled within the range of 0.3 to 0.5 meters.

[0066] After the partitioning is completed, the system synchronously collects wastewater samples from each layer using a undisturbed sampler to minimize disturbance or contamination of the samples during the collection process. The collected samples are then analyzed for chemical oxygen demand (COD) concentration using the standard potassium dichromate method. This method accurately quantifies the content of organic pollutants in wastewater through redox reactions, ensuring reliable results. The COD concentration data from each layer are recorded and processed by the system, and finally integrated to form a detailed depth distribution curve of COD concentration.

[0067] When dividing the wastewater pool into equal-distance layers in the equal-distance layering module, the system first accurately measures the total depth of the wastewater pool and, combined with the preset single-layer depth range of 0.3 to 0.5 meters, optimizes the layering strategy through rigorous mathematical calculations. Specifically, the total depth of the wastewater pool is divided by the lower limit of the single-layer depth of 0.3 meters and then rounded up to determine the minimum number of layers. To ensure that each layer has a depth of at least 0.3 meters to meet the minimum sampling requirement, the maximum number of layers is determined by dividing by the upper limit of single-layer depth of 0.5 meters and rounding down. To prevent a single layer depth from exceeding 0.5 meters and ensure processing efficiency; and The system intelligently selects the most suitable number of layers N based on the actual pool depth and operating conditions, enabling flexible adjustment.

[0068] After stratification, the system simultaneously collects wastewater samples from each stratum. The collected samples are then sent to the laboratory for repeated determination of chemical oxygen demand (COD) concentration using the standard potassium dichromate method. Each determination follows experimental procedures to eliminate random errors. Finally, the arithmetic mean of all results is calculated to generate high-precision, low-bias COD concentration data. This data is then integrated and visualized by the system to form a detailed depth distribution curve of COD concentration.

[0069] Establish a historical wastewater treatment database to obtain the number of historical independent electrolytic layers used in wastewater treatment ponds of the same depth. ;

[0070] like The total number of wastewater levels N at this point is related to the number of historical independent electrolysis layers. Same, that is ;

[0071] like Then the total number of wastewater levels N is the minimum value of the total number of levels. Same, that is ;

[0072] like The total number of wastewater levels N is the maximum value of the total number of levels. Same, that is .

[0073] Dynamic threshold zoning module: An initial COD concentration span threshold is set as the zoning benchmark. This threshold is pre-configured based on the wastewater pollution characteristics. Then, starting from the bottom of the wastewater tank and scanning layer by layer to the top, the system automatically calculates the COD concentration difference between two adjacent layers and compares it with the initial threshold in real time. When the COD concentration difference between adjacent layers is detected to be greater than or equal to the initial threshold, the system will immediately insert an independent electrolysis layer at the boundary to ensure that the high concentration difference area receives targeted treatment.

[0074] Conversely, if the concentration difference between adjacent layers is less than the initial threshold, the system will merge these two layers into the same electrolysis zone to avoid over-segmentation. After completing the traversal of all layers, the system automatically counts the total number M of the final generated independent electrolysis layers and accurately records the depth range coordinates covered by each electrolysis layer, such as the bottom boundary, top boundary, and range value.

[0075] In the dynamic threshold partitioning module, the formula for calculating the total number of electrolytic layers M is:

[0076]

[0077] in:

[0078] The total number of wastewater layers represents the total number of layers in the vertical direction of the wastewater tank. The larger N is, the more comparisons (N-1) are made between adjacent layers, and the more opportunities there are to trigger the conditions for inserting an independent electrolysis layer.

[0079] This is the numbering of the wastewater level, and The value range is [1, N-1];

[0080] For the first Layer and First The COD concentration difference between layers represents the difference in COD concentration between two adjacent layers. The greater the difference, the more likely it is to exceed the set initial COD concentration span threshold T;

[0081] This represents the threshold value for the initial COD concentration range;

[0082] and For indicator functions, when the condition The value is 1 when the condition is met, and 0 otherwise. It is used to quantify whether the difference in COD concentration between adjacent levels reaches the initial COD concentration span threshold. If this condition is not met, adjacent wastewater layers merge into the same electrolysis zone, and the M value does not increase.

[0083] The first Layer and First COD concentration difference between layers The calculation method is as follows:

[0084]

[0085] in:

[0086] For the first COD concentration of wastewater in the layer; For the first COD concentration of wastewater in the layer. It depends directly on the relative magnitude of COD concentrations between two adjacent layers. and The greater the difference, the better. The bigger

[0087] This reflects the uniformity of wastewater quality in the vertical direction. Greater differences indicate a more uneven vertical distribution of pollutants, potentially requiring more sophisticated stratified electrolysis treatment.

[0088] Multi-source coupling pretreatment module: The core task of the multi-source coupling pretreatment module is to collect multi-dimensional characteristic data closely related to the COD purification effect in real time. These data directly determine the efficiency and quality of electrolysis treatment. The specific categories collected include electrolysis time, current density, water pollution control agent concentration, conductivity and wastewater temperature.

[0089] Among these parameters, electrolysis duration reflects the impact of reaction duration on pollutant degradation, current density measures the energy input intensity of the electrolysis process, water pollution control agent concentration indicates the synergistic effect of chemical additives in purification, conductivity characterizes the ionic conductivity of wastewater, and wastewater temperature affects the reaction kinetic rate. All these characteristic data are collected synchronously and in real time through a high-precision sensor network deployed in the wastewater ponds, ensuring the timeliness and representativeness of the data.

[0090] After data collection, the system immediately performs normalization preprocessing, converting all feature values ​​to a standardized numerical range to eliminate biases caused by different units of measurement and facilitate subsequent integration and calculation. Simultaneously, to address missing values ​​that may occur during data collection, the system employs a sliding window mean imputation mechanism. This method constructs a dynamic window based on nearby time points or spatially adjacent valid data points, calculates the arithmetic mean of the data within the window, and intelligently imputes the missing values ​​to ensure dataset integrity.

[0091] In the multi-source coupling preprocessing module, the relevant feature data undergoes unified dimensional transformation and standardization. The maximum-minimum normalization method is adopted, which identifies the historical minimum and maximum values ​​of each feature data as reference points and linearly maps all feature values ​​to the standardized range of 0 to 1, thereby completely eliminating the problem of magnitude difference caused by different dimensionalities.

[0092] To address missing values ​​that may occur during data acquisition, the system has established an intelligent sliding window imputation mechanism. This mechanism flexibly utilizes the principles of time series continuity or spatial adjacency to automatically construct a dynamic window containing neighboring data points. The window size is intelligently adjusted to 4 data points based on the data acquisition frequency and distribution density to adapt to the data integrity requirements under different scenarios. Then, a reliable imputation value is generated by calculating the arithmetic mean of all valid data points within the window.

[0093] Based on long-term statistical data in the historical database, the system automatically calculates the arithmetic mean and standard deviation of each feature parameter as a benchmark. For real-time data points, threshold filtering is performed. When a value exceeds the range of the historical mean plus or minus three times the standard deviation, the system immediately identifies it as an outlier and performs intelligent removal, triggering a redundant data replacement process and utilizing a backup mechanism to compensate for missing data. The entire process is automated.

[0094] The redundant data replacement process includes:

[0095] The compensation mechanism activated when outliers or missing values ​​are identified is as follows: redundant datasets with the same parameter index are called. These datasets are derived from records of adjacent time nodes pre-stored by the system, real-time data collected from nearby spatial locations, or archived information from backup monitoring nodes. Then, the outlier data points are compared with the redundant datasets in detail to evaluate their correlation strength in the time and spatial dimensions, such as checking the synchronization trend or proximity consistency of the data.

[0096] Based on the comparison results, valid data points with high spatial correlation are prioritized as replacement candidates to ensure the high reliability of the replacement values. The strict condition for high spatial correlation is that within the same level of the wastewater tank, the radial distance between the candidate data point and the outlier does not exceed 0.2 meters. Candidate data that meet this condition will be automatically extracted and included in the replacement process, thereby seamlessly compensating for data loss and maintaining the stable operation of the entire wastewater treatment system.

[0097] If multiple valid candidate values ​​exist in the redundant dataset, a confidence-weighted fusion algorithm is used to generate the optimal replacement value. After the replacement is completed, the system immediately performs cross-parameter logical verification. If the verification fails, a second replacement is triggered. All replacement operations are marked with timestamps, original values, replacement values, and decision paths, and written to the audit log.

[0098] The formula for normalizing the relevant feature data in the multi-source coupling preprocessing module is as follows:

[0099]

[0100] in:

[0101] For the various relevant feature data collected; For normalized relevant feature data; This refers to the minimum baseline value for each relevant feature data; This represents the maximum baseline value for each relevant feature data;

[0102] During the normalization process, a corresponding static benchmark value is selected for each relevant characteristic data such as electrolysis time, current density, concentration of water pollution control agent, conductivity, and wastewater temperature. These benchmark values ​​are derived from the representative average values ​​accumulated over a long period of time after the wastewater has been left to stand in the historical database, ensuring the stability and reliability of the benchmark. The selection process involves automatically querying the database and extracting typical historical average values ​​of specific parameters as reference anchor points for normalization.

[0103] Establish an intelligent dynamic benchmark verification mechanism. This mechanism performs regular updates based on the distribution characteristics of historical data, such as mean drift or trend changes, and recalculates the benchmark value every quarter or after each batch of processing.

[0104] The normalized feature data is transformed into standardized variables with uniform dimensions. These variables are additive and can be directly used as input for subsequent weighted integration or optimization calculations.

[0105] The exponential-driven closed-loop optimization module calculates the electrolysis purification prediction index based on the preprocessed relevant feature data collected by the multi-source coupling preprocessing module, sets the electrolysis purification index threshold, and compares the electrolysis purification index threshold with the electrolysis purification prediction index to determine whether the final electrode layering quantity needs to be corrected.

[0106] In the index-driven closed-loop optimization module, the formula for calculating the electrolysis purification prediction index is as follows:

[0107]

[0108] in:

[0109] This is a predicted index for electrolysis purification.

[0110] The electrolysis time is the electrolysis time value after maximum-minimum normalization, ranging from [0,1]. A longer electrolysis time results in more complete redox reactions of pollutants on the electrode surface; electrolysis time is a key operating parameter affecting purification efficiency. Therefore, the weighting coefficients in this invention... Larger;

[0111] The current density is the value after maximum-minimum normalization, ranging from [0,1]. Current density directly affects the electrochemical reaction rate; increasing the current density can accelerate pollutant degradation. Its importance is second only to electrolysis duration but higher than temperature; therefore, it is a weighting factor in this invention. Above average;

[0112] The concentration of water pollution control agents is the agent concentration value after maximum-minimum normalization, ranging from [0,1]. Adding agents (such as coagulants and oxidants) can assist the electrolysis process. Since the agents are only used as an auxiliary means, the entire system focuses on electrolysis itself, therefore the weighting coefficient is... Minimum;

[0113] The conductivity value is the conductivity after maximum-minimum normalization, ranging from [0,1]. Conductivity reflects the conductivity of wastewater and determines the current transmission efficiency; conductivity is the fundamental guarantee for the electrolysis process, therefore the weighting coefficients in this invention... Larger;

[0114] The wastewater temperature is the temperature value after maximum-minimum normalization, ranging from [0,1]. Increased temperature can improve ion migration rate, enhance conductivity, and accelerate chemical reaction rates, but it can also lead to decreased electrode material stability and accelerated side reactions. Therefore, its importance is slightly higher than current density but lower than conductivity, and it is the weighting coefficient in this invention. medium;

[0115] These are the electrolysis time (t) and current density, respectively. Concentration of water pollution control agents Electrical conductivity The weighting coefficients for wastewater temperature K and the total number of electrolytic layers M are determined, and each weighting coefficient satisfies the following relationship: ,and .

[0116] The threshold for the electrolytic purification index is set as follows: The calculated electrolysis purification prediction index With the threshold of the electrolytic purification index Perform real-time comparison:

[0117] like If the total number of electrolytic layers M and the corresponding relevant characteristic data meet the purification requirements, the process will directly enter the execution stage.

[0118] like This indicates that the current electrolytic purification effect has failed to meet expectations, especially when dealing with high-concentration water pollution. At this point, an adaptive correction mechanism is triggered—the system dynamically reduces the COD concentration range threshold. In this embodiment, the concentration is decreased in increments of 0.05 mg / L, based on the updated COD concentration range threshold. The hierarchical logic in the dynamic threshold partitioning module is re-executed to generate a larger total number of electrolysis layers. As the total number of electrolysis layers increases, the system automatically adjusts the key characteristic data parameters corresponding to each electrolysis layer, such as the current density or processing time allocated to each layer. After updating the parameters and structure, the system performs a comprehensive calculation to obtain the electrolysis purification prediction index under the new configuration parameters. It then recalculates the electrolysis purification prediction index under the updated parameters and compares it a second time with the electrolysis purification index threshold. If the comparison result still does not meet the threshold requirement, the entire adaptive correction process (including adjusting the threshold T, re-layering, updating parameters, recalculating, and comparing) is iteratively executed until the electrolysis purification index threshold requirement is reached or exceeded, thus ensuring that the purification effect remains stable even in dynamically changing environments. Ultimately, this supports the overall goal of wastewater treatment and its reuse.

[0119] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0120] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0121] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0122] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A system for treating wastewater from pickling vegetables, characterized in that, Includes the following modules: Equidistant stratification module: The pickling wastewater pool is divided into equal intervals according to vertical depth. After the wastewater is allowed to settle, N continuous wastewater levels are generated. The depth range of each level is [0.3-0.5] meters. Wastewater samples are collected from each level simultaneously, and the chemical oxygen demand (COD) concentration of each level is determined by the standard potassium dichromate method to form a depth distribution curve of COD concentration. Dynamic threshold partitioning module: Set an initial COD concentration span threshold, calculate the COD concentration difference between adjacent layers from the bottom to the top; when the COD concentration difference between adjacent layers is not less than the initial COD concentration span threshold, an independent electrolysis layer is inserted; when the COD concentration difference between adjacent layers is less than the initial COD concentration span threshold, they are merged into the same electrolysis zone; after traversing all layers, count the total number M of independent electrolysis layers, and record the depth range covered by each electrolysis layer. Multi-source coupling preprocessing module: Collects relevant feature data directly related to COD purification effect. The relevant feature data includes electrolysis time, current density, water pollution control agent concentration, conductivity and wastewater temperature. All relevant feature data are normalized preprocessed, and missing values ​​are filled with the mean of a sliding window. The index-driven closed-loop optimization module calculates the electrolytic purification prediction index based on the preprocessed relevant feature data collected by the multi-source coupling preprocessing module, sets the electrolytic purification index threshold, compares the electrolytic purification index threshold with the electrolytic purification prediction index, and determines whether the initial COD concentration span threshold needs to be corrected.

2. The treatment system for pickled vegetable wastewater according to claim 1, characterized in that: When performing equidistant division in the equidistant layering module, the minimum number of layers is determined by calculating the pool depth divided by the lower limit of the single-layer depth (0.3 meters) and rounding up, based on the total depth of the wastewater pool and the preset single-layer depth range. At the same time, divide by the upper limit of single-layer depth of 0.5 meters and round down to determine the maximum number of layers. Finally, the number of layers was selected between the minimum and maximum values. When collecting wastewater samples from each layer simultaneously, the chemical oxygen demand (COD) concentration was measured multiple times using the standard potassium dichromate method. The average value was then used to generate a high-precision COD concentration depth distribution curve.

3. The treatment system for pickled vegetable wastewater according to claim 2, characterized in that: Establish a historical wastewater treatment database to obtain the number of historical independent electrolytic layers used in wastewater treatment ponds of the same depth. ; like The total number of wastewater levels N at this point is related to the number of historical independent electrolysis layers. Same, that is ; like Then the total number of wastewater levels N is the minimum value of the total number of levels. Same, that is ; like The total number of wastewater levels N is the maximum value of the total number of levels. Same, that is .

4. The treatment system for pickled vegetable wastewater according to claim 2, characterized in that: In the dynamic threshold partitioning module, the formula for calculating the total number of electrolytic layers M is: in: This represents the total number of wastewater treatment levels. This is the numbering of the wastewater level, and The value range is [1, N-1]; For the first Layer and First COD concentration difference between layers; This represents the threshold value for the initial COD concentration range; and For indicator functions, when the condition The value is 1 when the condition is met, and 0 otherwise. It is used to quantify whether the difference in COD concentration between adjacent levels reaches the initial COD concentration span threshold. If this condition is not met, adjacent wastewater layers merge into the same electrolysis zone, and the M value does not increase.

5. The treatment system for pickled vegetable wastewater according to claim 4, characterized in that: The first Layer and First COD concentration difference between layers The calculation method is as follows: in: For the first COD concentration of wastewater in the layer; For the first COD concentration of wastewater in the layer.

6. The system for treating pickled vegetable wastewater according to claim 1, characterized in that: In the multi-source coupling preprocessing module, the relevant feature data undergoes unified dimensional transformation and standardization, and the maximum-minimum normalization method is used to map each feature value to the [0,1] interval to eliminate the influence of magnitude differences on subsequent calculations; To address missing values ​​that may occur during data collection, a sliding window mechanism will be established based on time series or spatial adjacency. The missing values ​​will be filled by calculating the arithmetic mean of the valid data within the window, and the window size will be dynamically adjusted to 3-5 data points. The system performs threshold filtering on outliers, removing those that exceed the historical data mean by ±3 times the standard deviation, and initiating a redundant data replacement process.

7. The treatment system for pickled vegetable wastewater according to claim 6, characterized in that, The redundant data replacement process includes: The compensation mechanism activated when outliers or missing values ​​are identified is as follows: Relevant redundant datasets of the same related feature data are collected at adjacent time points, spatial locations, or backup monitoring nodes. By comparing the spatiotemporal correlation between outlier data points and redundant datasets, valid relevant redundant data with high spatial correlation are selected as replacement candidates. The condition for high spatial correlation is: relevant feature data within a radial distance of ≤0.2 meters on the same layer. If multiple valid candidate values ​​exist in the relevant feature redundancy dataset, a confidence-weighted fusion algorithm is used to generate the optimal replacement value. After the replacement is completed, the system immediately performs cross-parameter logical verification. If the verification fails, a second replacement is triggered. All replacement operations are marked with timestamps, original values, replacement values, and decision paths, and written to the audit log.

8. The treatment system for pickled vegetable wastewater according to claim 7, characterized in that: The formula for normalizing the relevant feature data in the multi-source coupling preprocessing module is as follows: in: For the various relevant feature data collected; For normalized relevant feature data; This refers to the minimum baseline value for each relevant feature data; This represents the maximum baseline value for each relevant feature data; During the normalization process, a corresponding static benchmark value is selected for each relevant feature data. The static benchmark value is the representative average value of wastewater after settling in the historical database. A dynamic verification mechanism for the benchmark value is established, and the benchmark value is updated periodically based on the distribution characteristics of historical data. The normalized relevant feature data are superimposed input variables.

9. The system for treating pickled vegetable wastewater according to claim 5, characterized in that, In the index-driven closed-loop optimization module, the formula for calculating the electrolysis purification prediction index is as follows: in: This is a predicted index for electrolysis purification. The electrolysis time is the electrolysis time value after maximum-min normalization, ranging from [0,1]. is the current density, the current density value after maximum-min normalization, ranging from [0,1]. The concentration of water pollution control agent is the agent concentration value after maximum-minimum normalization, ranging from [0,1]. represents the conductivity, a conductivity value after maximum-minimum normalization, ranging from [0,1]. The wastewater temperature is the temperature value after maximum-minimum normalization, ranging from [0,1]. These are the electrolysis time (t) and current density, respectively. Concentration of water pollution control agents Electrical conductivity The weighting coefficients for wastewater temperature K and the total number of electrolytic layers M are determined, and each weighting coefficient satisfies the following relationship: ,and .

10. A treatment system for pickled vegetable wastewater according to claim 9, characterized in that: The threshold for the electrolytic purification index is set as follows: The calculated electrolysis purification prediction index With the threshold of the electrolytic purification index Perform real-time comparison: like If the total number of electrolytic layers M and the corresponding relevant characteristic data meet the purification requirements, the process will directly enter the execution stage. like This triggers an adaptive correction mechanism—the system dynamically reduces the COD concentration range threshold. Based on the updated COD concentration span threshold The hierarchical logic in the dynamic threshold partitioning module is re-executed to generate a larger total number of electrolysis layers. At the same time, the relevant feature data of each electrolysis layer are adjusted in conjunction with the update parameters. Then, the electrolysis purification prediction index under the updated parameters is recalculated and compared with the electrolysis purification index threshold for the second time until the electrolysis purification index threshold requirement is reached or exceeded.

Images